I studied content delivery in distributed systems with highly heterogeneous resources. I aimed at designing and developing new adaptive and autonomous solutions for video delivery over the Internet advocating for the simultaneous usage of multiple sources with heterogeneous capacities so as to achieve high QoE content delivery at low cost.
My research contributions can be summarised as follows:
Adaptive multiple-source streaming over HTTP for enhancing end-users’ quality of experience
I studied, the designed and the developed a solution to enable the support of Multiple-Source STREAMing (MS-Stream) in current widely deployed HTTP Adaptive Streaming techniques for enhancing the Quality of Experience of end-users when consuming video delivery services.
Multi-criteria optimization for the content delivery process in multiple-server streaming
In order to cope with the resource heterogeneity in multi-server environment, I proposed an adaptive multi-criteria optimization of the video delivery process in the above presented MS-Stream solution. The latter optimization aims at a reaching a predefined Service Level Objective to satisfy the end-users when consuming content.
P2P/multi-server adaptive live video delivery
I designed and developed a P2P/Multi-Server video streaming system proposal relying on MS-Stream to provide the highest possible quality of experience and to optimize the resources made available at the server infrastructure. I introduced a first local optimization method adapt the video quality consumed by end-users according to the global and local « health » of the distributed system, and providing the best possible experience to the maximum number of consumers. I proposed a second local optimization method for minimizing the resource utilization of the fixed-size server infrastructure made by the peers without sacrificing the achieved quality of experience gains.